Mahfuz Karim

We have battled with the issue of defining 'good dispersion' in ISO TC24/SC4 Particle Characterisation without any real formal outcome other than generalities.

A lot depends on the end application plus the size distribution, in particular the number of large particles (sparse sampling). 'Good' and 'poor' are subjective terms as you know.

Only route in a matrix that I would think viable would be:

• Calculate the best standard error possible based on the particle size distribution and some other properties. See:

Keys for Successful Analysis – Representative Sampling & Estimation of Standard Error Calculation

https://www.malvernpanalytical.com/en/learn/events-and-training/webinars/W190613KeysLD.html

• Map out a number of (random) areas on your matrix (e.g. 1 cm2) and calculate the key size parameters (e.g. x10, x50, and x90) and carry out a chi-squared statistical test to indicate if there is significant differences between the mapped areas. If there is then this may suggest a process such as segregation or over-specification (especially on the x90+), noting that one would need 10000 particles to specify the x99 to 1% standard error

BTW, how are Brunel and Uxbridge? I was there from 1972 to 1979...

Surface morphology analysis by Scanning electron microscope will exhibit good dispersion surface as a quite smooth surface with more uniformity and enhanced compactness with good filling effect of the existing pores.

Poor distribution is depicted by surface roughness and agglomeration and aggregation of particles

Preethi Shetty

OK, then define what metrics and instrumentation we can use to define 'quite smooth surface with more uniformity and enhanced compactness with good filling effect of the existing pores'...

Sir surface finish can be assessed by profilometer which can measures the roughness of a surface

Riemann metric

Preethi Shetty

OK, so with your Riemann plot (log-log), then what is ‘good’ and what is ‘bad’? What if the Riemann plot is not linear? What if the particles are in the matrix but not raised - for example, a polished specimen? The question was ‘within a matrix’ not ‘on a matrix’...

Alan F Rawle Thanks for the advice. I will give it a go. My plan was to develop a system where I can denote my different sample with label on the distribution to make it easier for me in my Thesis.

I am also looking to get CT scans of the sample to see the spread of particles within the sample, but creating a 3d representative of the distribution would be tricky.

Brunel is doing well, everyone is doing the best we can during the lockdown period.